W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > November 2001

Re: Resolving rdfms-assertion: media type draft, IRC discussion

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Fri, 16 Nov 2001 17:49:58 -0600
Message-Id: <p05101084b81b53324548@[65.212.118.147]>
To: Aaron Swartz <me@aaronsw.com>
Cc: w3c-rdfcore-wg@w3.org
>Well, I just finished reading an interesting discussion on #rdfig on 
>this issue[1], and folks seem to be making it more complicated than 
>I think it is, but I hope to provide some ideas on where to go. 
>Here's what my Media Type draft says:
>
>"""
>Because RDF is a format for semantically-meaningful information, it 
>is important to note that transmission of RDF under this media type, 
>whether via HTTP, SMTP or some similar protocol, means that the 
>sender asserts the content of the RDF document.

Might be better to say, the sender intends to assert the content of....

But the main problem with this is that 'the content of' isn't clear 
for RDF, or indeed for any assertional formal language. Suppose I 
were to publish some FOL axioms, for example, which tie their 
meanings down a lot more tightly than RDF does; what would it mean to 
say that I was asserting their content? They have no particular 
content in the real world; it depends on how one interprets the 
symbols that occur in them.

Maybe what we need to say here is that the content of any published 
RDF is intended to take into account all agreed meanings of any URIs 
that occur in the RDF, and that such meanings may be specified by 
communal, social or legal agreements made externally to RDF itself; 
and that the occurrence of any URI in any asserted RDF shall be 
deemed to constitute a public use of that URI (or whatever the 
appropriate legal phrase is: is there a lawyer in the house?), and 
that any conclusions involving such usage that are validly entailed 
by any published RDF shall be understood to have the same legal 
force, and to incur all obligations and rights accruing from said 
publication, that they would if they had been included explicitly in 
the publication.

>If this is not desirable, such as when a system is forwarding RDF 
>written by someone else, another applicable media type, such as 
>application/xml or application/octet-stream should be used. Also 
>note that RDF provides reification so that RDF statements can be 
>sent and discussed without actually being asserted themselves.
>"""
>  - http://blogspace.com/rdf/mimetype
>    Media Type for Resource Description Framework (RDF)
>
>Tim Berners-Lee brought up three points which he felt would close this issue:
>
>1) The meaning of a document is the conjunction of the meanings of 
>its statements.

Almost true (not quite exactly when bNodes occur), but unhelpful.

>2) The meaning of a statement is defined by the specification of its 
>predicate.

Definitely not true, in general.

>3) The meaning of { :x rdf:type :c} is defined by the rdf:type 
>definition which delegates the meaning to the definition of :c .

? Incomprehensible. The meaning of {:x rdf:type :c} is that :x is in 
the class :c.

Pat
-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Friday, 16 November 2001 18:49:58 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:42:45 EDT